Testing a Quantity Break Discount Table on Product Pages

Quantity break tables, buy two save ten percent, buy three save twenty, are a fixture on consumable and multi buy product pages. They can lift units per order in a way few other elements can, and they can also train shoppers to expect discounts, clutter the buying area, and quietly hand margin to people who would have bought anyway. That makes them a proper lab subject: a test with a margin guardrail, not a plugin installed on a hunch. This design tests the table on one product group, against a clean control, with success defined before the first order arrives.

Choose the right products and break points from your data

Quantity breaks work where buying more is natural: consumables, refills, gifts bought in multiples, and supplies. They fail on considered, one off purchases where a second unit has no use. Pick one product group with healthy repeat or multi unit behavior already visible in your orders, because the table amplifies existing behavior more than it creates new behavior. Set break points from your current distribution: if most orders are one unit and a meaningful minority are two, the first break belongs at two, priced so the discount is funded by shipping and handling savings on the combined order, not by hope. Write the margin per scenario down before the test starts.

Present the table as one clear variable

The test version shows a compact table or tier list near the buying controls: quantities, per unit price at each tier, and the saving in plain numbers. The control shows the standard single price with no table. Keep everything else identical: same images, same reviews position, same shipping message. Pre select nothing aggressive; defaulting the quantity to a discounted tier without the shopper choosing it confuses the test and annoys shoppers. Make sure the cart and checkout reflect the tier pricing exactly as the table promised, because a mismatch discovered at checkout does not just lose the sale, it teaches shoppers your tables lie.

Measure units, margin, and hesitation together

Primary metric: units per order on the tested products. Guardrails: contribution margin per order on those products, conversion rate, and discount related support questions. A table that lifts units twenty percent while margin per order falls has not won; it has bought revenue at a loss. Run through at least two full weekly cycles and avoid promotional weeks, since sitewide sales stack on top of tier discounts in ways that make both unreadable. Segment new versus returning shoppers if your tool allows, because returning buyers often use breaks as a stock up mechanism, which is the healthy version of the behavior.

Decide by segment, and document the pricing logic

If units and margin per order both rise, ship the table for that product group and consider a follow up test on presentation, such as highlighting the most popular tier. If units rise and margin falls, try shallower discounts at the same break points as a second test before abandoning the idea. If nothing moves, the product group was probably wrong, not the concept. Whatever the outcome, record the break points, the discount math, and the result in your experiment log with dates. Quantity pricing has a long memory with customers; changing it casually every quarter trains waiting behavior that no test can fix.

Frequently asked questions

Where should the quantity break table sit on the page?

Near the quantity selector and add to cart controls, where the shopper is deciding how many to buy. A table buried below the description arrives after the decision. Keep it compact: tiers, per unit prices, and savings, readable at a glance on mobile. Highlighting the tier most shoppers choose can help, but test that as a separate change later. During the test, the table placement is part of the single variable, so resist adding banners and badges promoting it elsewhere on the page.

How deep should quantity discounts be?

Deep enough to be worth the shopper planning, shallow enough to be funded by real savings: combined shipping, one pick and pack instead of two, and lower service cost per unit. Many stores land between five and twenty percent across tiers. Model each tier margin before the test and treat those numbers as guardrails during it. If a tier only makes sense assuming shoppers would not have bought otherwise, your own order data from the control group will tell you whether that assumption survived contact with reality.

Will quantity breaks train shoppers to never pay full price?

On the tested products, some of that effect is real and acceptable if the multi unit behavior is profitable and natural, such as genuine stocking up. The risk grows when breaks spread to every product, including one off purchases where they only discount intent that already existed. Keep breaks scoped to product groups where multiples make sense, keep the table factual rather than promotional in tone, and review the full price share of orders in your quarterly results. Scope is the control on this risk.

Can I run a quantity break test during a sale period?

Avoid it. Sitewide promotions interact with tier pricing: stacking rules decide which discount wins, shoppers behave differently under sale urgency, and the results will describe the sale, not the table. If a sale is unavoidable mid test because of the calendar, note it, and extend the test to include a clean comparison window. The general lab rule applies: one major change at a time, and outside events that change buying behavior count as changes, even when they are not yours.